beautiful planet earth glowing as the shell of a giant turtle floating in space
beautiful planet earth glowing as the shell of a giant turtle floating in space

news

AI-generated Drake and The Weeknd song goes viral (bbc.com) | [💬 discussion]- A song that uses Artificial Intelligence to clone the voices of Drake and The Weeknd has gone viral on social media.

RedPajama: Reproduction of Llama with Friendly License (together.xyz) | [💬 discussion] - RedPajama, a project to create leading open-source models, starts by reproducing LLaMA training dataset of over 1.2 trillion tokens

OpenAI’s CEO Says the Age of Giant AI Models Is Already Over (wired.com) | [💬 discussion] - Sam Altman says the research strategy that birthed ChatGPT is played out and future strides in artificial intelligence will require new ideas.

Community is the future of AI (stackoverflow.blog) | [💬 discussion]- Stack Overflow CEO to appeal to the community in the wake of description of their unique selling point.

Superchat’s new AI chatbot lets you message historical and fictional characters via ChatGPT (techcrunch.com) - The company behind the popular iPhone customization app Brass, sticker maker StickerHub and others is out today with a new AI chat app called Superchat, which allows iOS users to chat with virtual characters powered by OpenAI’s ChatGPT. However, what makes the app different from the default ChatGPT experience or the dozens of generic AI chat apps now available are the characters offered, which you can use to engage with Superchat’s AI features.

Synthetic Data from Diffusion Models Improves ImageNet Classification (arxiv.org) | [💬 discussion] - Deep generative models are becoming increasingly powerful, now generating diverse high fidelity photo-realistic samples given text prompts. Have they reached the point where models of natural images can be used for generative data augmentation, helping to improve challenging discriminative tasks? We show that large-scale text-to image diffusion models can be fine-tuned to produce class conditional models with SOTA FID (1.76 at 256x256 resolution) and Inception Score (239 at 256x256). The model also yields a new SOTA in Classification Accuracy Scores (64.96 for 256x256 generative samples, improving to 69.24 for 1024x1024 samples). Augmenting the ImageNet training set with samples from the resulting models yields significant improvements in ImageNet classification accuracy over strong ResNet and Vision Transformer baselines.

GPT-4 will hunt for trends in medical records thanks to Microsoft and Epic (arstechnica.com) - Generative AI promises to streamline health care, but critics say not so fast.

Elon Musk wants to develop TruthGPT (techcrunch.com) - he wants to develop his own chatbot called TruthGPT, which will be “a maximum truth-seeking AI”.

Reddit will begin charging for access to its API (nytimes.com) | [💬 discussion]- Reddit Wants to Get Paid for Helping to Teach Big A.I. Systems. The internet site has long been a forum for discussion on a huge variety of topics, and companies like Google and OpenAI have been using it in their A.I. projects.

Microsoft’s AI Chip Strategy Reduces Costs and Nvidia Dependence (artisana.ai) - Microsoft’s own AI chip, codenamed Project Athena, could help cut their ballooning AI costs by up to a third. With ChatGPT costing OpenAI an estimated $700,000 per day to run, computing expenses are top of mind for many AI companies.

Adobe teases generative AI video tools (arstechnica.com) - “Firefly for Video” will generate sound effects and music, edit footage via text.

🐦 Top 10 Twitter thread of AutoGPT (twitter.com) - AutoGPT has EXPLODED in the past 5 days. Reaching #1 trending on Twitter. It’s the latest & greatest AI you’ve seen yet.

NaturalSpeech 2 (speechresearch.github.io) - Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers

Stability AI Launches the First of its StableLM Suite of Language Models (stability.ai) | [💬 discussion]- Today, Stability AI released a new open-source language model, StableLM. The Alpha version of the model is available in 3 billion and 7 billion parameters, with 15 billion to 65 billion parameter models to follow. Developers can freely inspect, use, and adapt our StableLM base models for commercial or research purposes, subject to the terms of the CC BY-SA-4.0 license.

The secret list of websites that make AI chatbots sound smart (washingtonpost.com) | [💬 discussion] - Inside the secret list of websites that make AI like ChatGPT sound smart

Record labels panic as A.I. generated song becomes ‘hit single’ (darkfutura.substack.com) | [:speech_balloon discussion] - An interesting landmark case has turned over a new milestone in AI development this week. The first AI-generated ‘hit single’ has appeared. A song featuring AI synthesized versions of hip-hop/R&B artists Drake and The Weeknd has garnered over 15 million views, with many fans calling the song far better than anything from the recent ouvre of either human artist. Universal Records panicked and called on their legal clout to have the song pulled off from streaming services.

Google consolidates AI research divisions into Google DeepMind (techcrunch.com) - As Google looks to maintain pace in AI with the rest of the tech giants, it’s consolidating its AI research divisions.

ChatGPT is taking ghostwriters’ jobs in Kenya - Rest of World (restofworld.org) | [💬 discussion] - Kenya is a hub for freelancers who help American students write essays and do homework. The rise of AI tools like ChatGPT is reducing their earnings.

How Much Does ChatGPT Cost to Run? $700K/day, Per Analyst (businessinsider.com) | [💬 discussion] - ChatGPT is costly because it requires massive amounts of computing power to generate answers to queries. Microsoft is making a chip to change that.

Large, creative AI models will transform lives and labour markets | The Economist (economist.com) | [💬 discussion] - They bring enormous promise and peril. In the first of three special articles we explain how they work

GitHub Copilot Emits GPL. Codeium Does Not. | Codeium (codeium.com) | [💬 discussion] - Demonstrating that GitHub Copilot trains on non-permissive licenses and is unable to filter out suggestions properly, while Codeium does not expose users to legal risk.

‘I’ve Never Hired A Writer Better Than ChatGPT’: How AI Is Upending The Freelance World (forbes.com) | [💬 discussion] - While some freelancers are losing their gigs to ChatGPT, clients are being spammed with AI-written content on freelancing platforms. The result: increasing mistrust between clients and freelancers and mounting trouble for the platforms themselves.

This new technology could blow away GPT-4 and everything like it | ZDNET (zdnet.com) | [💬 discussion] - The Hyena code is able to handle amounts of data that make GPT-style technology run out of memory and fail.

There Is No A.I. | The New Yorker (newyorker.com) | [💬 discussion] - There are ways of controlling the new technology—but first we have to stop mythologizing it.

Project Magi: Google’s team of 160 working on adding new features to search engine - Tech (me.mashable.com) - Google is also developing a new search engine that promises to deliver a more personalized experience for users.

GrammarlyGO | Grammarly (grammarly.com) - Introducing GrammarlyGO, the suite of generative AI capabilities from the leader in AI communication assistance.

tool

MiniGPT-4 (minigpt-4.github.io) | [💬 discussion] - Enhancing Vision-language Understanding with Advanced Large Language Models

Superpower ChatGPT (chrome.google.com) - ChatGPT with superpowers! Sync/search history locally, create folders, export all chats, pin messages, access thousands of prompts.

HeyGPT (heygpt.chat) - Get more out of ChatGPT. Use your own API key. No login required. Access ChatGPT anytime, including peak periods. Faster response time. Unlock advanced functionalities such as chat with multiple pdfs at once, using ChatGPT with text-to-speech, speech recognition, chat search, tagging your chats and more.

cligpt (github.com) - Streamline your terminal experience by generating Linux commands from natural language queries, reducing the need to leave the terminal for manual web searches.

text-generation-webui (github.com) - A gradio web UI for running Large Language Models like LLaMA, llama.cpp, GPT-J, Pythia, OPT, and GALACTICA. Its goal is to become the stable-diffusion-webui of text generation.

promptr (github.com) - About Promptr is a CLI tool that makes it easy to apply GPT’s code change recommendations with a single command. With Promptr, you can quickly refactor code, implement classes to pass tests, and experiment with LLMs. No more copying code from the ChatGPT window into your editor.

awesome-chatgpt-prompts (github.com) - This repo includes ChatGPT prompt curation to use ChatGPT better.

Single Prompt AI (singlepromptai.com) - A collection of single purpose AI interfaces. Each of these AI do one thing and one thing only.

AI Playground by Vercel Labs (play.vercel.ai) - Compare, tune, and play with the latest AI language models.

AI app that auto-creates presentations from just a topic (magicslides.app) - Ideas 💡 to Professional Presentations in Seconds ⏱️

Checksum (checksum.ai) - Test every corner of your app using AI. Checksum automatically generates and maintains end-to-end tests based on user sessions so you can move fast without breaking things.

MeetGeek (meetgeek.ai) - AI-generated summary for your Zoom, Teams & GMeet calls

openpm (openpm.ai) - The OpenAPI package manager

bark (github.com) - Bark is a transformer-based text-to-audio model created by Suno. Bark can generate highly realistic, multilingual speech as well as other audio - including music, background noise and simple sound effects. The model can also produce nonverbal communications like laughing, sighing and crying. To support the research community, we are providing access to pretrained model checkpoints ready for inference.

SafeGPT - Avoid errors & biases in ChatGPT (giskard.ai) | [💬 discussion] - Check for errors, biases & privacy issues in ChatGPT & other LLMs | Get early access to SafeGPT, our browser extension & quality assurance platform.

WasmGPT (lxe.co) | [💬 discussion]

GitHub - sanchit-gandhi/whisper-jax (github.com) - Contribute to sanchit-gandhi/whisper-jax development by creating an account on GitHub.

Open Assistant (open-assistant.io) - Conversational AI for everyone. In the same way that Stable Diffusion helped the world make art and images in new ways, we want to improve the world by providing amazing conversational AI.

dup.ai (dup.ai) - AI-powered 3D editor designed for all, the only limit is your imagination. Seamlessly connecting the physical and digital realms, making 3D accessible and intuitive for all.

Prompting Guide (promptingguide.ai) - Prompt engineering is a relatively new discipline for developing and optimizing prompts to efficiently use language models (LMs) for a wide variety of applications and research topics. Prompt engineering skills help to better understand the capabilities and limitations of large language models (LLMs).

misc

Trivai.app (trivai.app) - AI powered trivia questions, with references

MemeCam (memecam.io) - Upload a picture or take one using your camera and let AI do its magic

Plask (plask.ai) - Plask is a web-based AI solution for generating 2D images with tailored poses and angles, streamlining the creation of captivating visuals for characters.

ReRoom AI (reroom.ai) - Fresh Ideas for Interior Design. Upload a picture of your room, explore various design styles(20+), and remodel your space today!

AutoGPT by SamurAI (autogpt.thesamur.ai) - We are building a free frontend for ChatGPT api with prompt library and chat history support

One Sentence Prompt (oneusefulthing.org) - Prompting for maximum impact (and why that is a bad idea)

An example of LLM prompting for programming (martinfowler.com) | [💬 discussion] - My account of an internal chat with Xu Hao, where he shows how he drives ChatGPT to produce useful self-tested code. His initial prompt primes the LLM with an implementation strategy (chain of thought prompting). His prompt also asks for an implementation plan rather than code (general knowledge prompting). Once he has the plan he uses it to refine the implementation and generate useful sections of code.

🎬 The A.I. Dilemma (youtube.com) | [📝 text-summary] - Tristan Harris and Aza Raskin discuss how existing A.I. capabilities already pose catastrophic risks to a functional society, how A.I. companies are caught in a race to deploy as quickly as possible without adequate safety measures, and what it would mean to upgrade our institutions to a post-A.I. world.

AI Incident Database (incidentdatabase.ai) - AI Incident Database

The Anatomy of Autonomy: Why Agents Are the Next AI Killer App (latent.space) - Auto-GPT/BabyAGI Executive Summary, a Brief History of Autonomous Agentic AI, and Predictions for Autonomous Future

How to train your own Large Language Models (blog.replit.com) - How Replit trains Large Language Models (LLMs) using Databricks, Hugging Face, and MosaicML

Autonomous Agents & Agent Simulations (blog.langchain.dev) | [💬 discussion] - Over the past two weeks, there has been a massive increase in using LLMs in an agentic manner. Specifically, projects like AutoGPT, BabyAGI, CAMEL, and Generative Agents have popped up. The LangChain community has now implemented some parts of all of those projects in the LangChain framework. While researching and implementing these projects, we’ve tried to best understand what the differences between them are and what the novel features of each are. This blog is an explanation of what we’ve learned.

🎬 The inside story of ChatGPT’s astonishing potential (ted.com) | [💬 discussion] - In a talk from the cutting edge of technology, OpenAI cofounder Greg Brockman explores the underlying design principles of ChatGPT and demos some mind-blowing, unreleased plug-ins for the chatbot that sent shockwaves across the world. After the talk, head of TED Chris Anderson joins Brockman to dig into the timeline of ChatGPT’s development and get Brockman’s take on the risks, raised by many in the tech industry and beyond, of releasing such a powerful tool into the world.

Building A ChatGPT-enhanced Python REPL (isthisit.nz) - In this blog I share my experience in building a Python REPL augmented with ChatGPT. I explore how the application is built, and speculate on software engineering patterns and paradigms that might emerge in systems built on Large Language Models (LLMs). GEPL - Generate, Evaluate, Print, Loop Link to this section Introduction The Lisp programming language made REPLs (Read, Evaluate, Print, Loop) famous. REPLs are interactive programming environments where the programmer gets immediate feedback on lines of code they just typed.

AI’s Perspective Hierarchy - A ShareGPT conversation (sharegpt.com) - Explain the human perspective hierarchy in the context of CEOs at the top of the perspective hierarchy and individual contributors being at the bottom using clear and precise expressions of concepts and ideas.

Prompt Engineering vs. Blind Prompting – Mitchell Hashimoto (mitchellh.com) | [💬 discussion] - In this blog post, I will make the argument that prompt engineering is a real skill that can be developed based on real experimental methodologies. I will use a realistic example to walk through the process of prompt engineering a solution to a problem that provides practical value to an application.

All Prompt Libraries in One Page (@samir) | Hero (hero.page) - all prompt libraries in one page created by @samir on Hero. Hero is a link sharing platform that allows you to easily share the amazing links you’ve discovered!

🐦 view tweet (twitter.com) - I’m a lawyer. I use ChatGPT daily. I feel like causing some chaos in my profession.

🐦 view tweet (twitter.com) - I connected ChatGPT to my personal health data on my iPhone. Now, I can have a conversation with my digital health history. The code is also public. Say hello to HealthGPT 👋

🐦 view tweet (twitter.com) - Excited to introduce NexusGPT, a marketplace of 800+ autonomous AI-agent with each specific skillset. Demo is free on http://nexus.snikpic.io

🐦 view tweet (twitter.com) - Today we’re launching Forefront chat—a better ChatGPT experience—in free alpha. Sign up to get free access to GPT-4, image generation, custom personas, shareable chats, and much more: http://chat.forefront.ai

The History of AI in 7 Experiments - by Mario Gabriele (thegeneralist.substack.com) - The breakthroughs, surprises, and failures that brought us to today.

⬡ ⬡ ⬡ ⏣ ⬡ ⬡ ⬡

share

hackernewslobstersreddittwitter